Childhood anxiety disorders (AD) are the most common mental health disorder in youth, with lifetime prevalence rates exceeding 30%. Research shows that childhood AD lead to significant and chronic impairment, including school underachievement and early dropout; peer difficulties; physical health problems; depression and substance use; and increased risk for suicide. Consequently, there is a strong need to identify potential mechanisms involved in the development of childhood AD, from public health, prevention, and economic standpoints. Given the central role of information processing factors in childhood AD, as well as the importance of the parent-child relationship in shaping how children interpret their environment, researchers have highlighted the need for research examining how parents shape child interpretation strategies known to be causally related to childhood anxiety. More specifically, research is needed to examine if maternal interpretation biases heighten the risk for childhood anxiety by increasing child interpretation biases. The goal of this study is to examine the effects of maternal interpretation biases on child interpretation biases and child anxiety symptoms among 130 children with AD. Clinically anxious mothers of children with AD will be randomly assigned to either a training program that teaches them to interpret child-related ambiguous situations in a positive direction or a neutral condition that does not train mothers to interpret situations in either a positive or negative manner. Children will subsequently be asked to participate in an anxiety-provoking speech task. Mothers and their children will discuss how they will approach the task; mothers' behavior during this discussion will be coded by observers blind to training condition. Following the discussion, children and mothers interpretation biases will be assessed. Finally, children's anxious behavior, physiological arousal, and performance during the task will be assessed. We expect that clinically anxious mothers who are trained to interpret child-related ambiguous situations in a positive manner (vs. those who do not receive such training) will exhibit reduced anxious behavior and rate the speech task as less threatening. We also expect that children of clinically anxious mothers who are trained to interpret child-related ambiguous situations in a positive manner will rate the speech task as less threatening than children of clinically anxious mothers who do not receive interpretation bias training. We also predict that children of clinically anxiou mothers who receive positive interpretation bias training (vs. those who do not receive such training) will exhibit lower anxiety during the speech task, as evidenced by (a) lower observer-rated anxious behavior, (b) increased respiratory sinus arrhythmia, and (c) higher observer-rated performance during the speech task. Finally, we predict that a social phobia (SoP) diagnosis in children will moderate the association between maternal interpretation biases and child anxiety symptoms, such that the association between maternal interpretation biases and child anxiety during the speech task will be stronger for children with SoP vs. other AD.